Titin.net

In previous posts, I illustrated the benefits of having consistent interfaces. Consistent interfaces uses our brains’ dual information processing capacity. In my tacit knowing post, I presented need for care for the smallest details in interface design. This post is the how. How should we design these interfaces? Do we need domain specific guides? Two examples of healthcare interface guides are Microsoft Healthcare Common User Interface (MSCUI) guide being developed with UK’s National Health Services (NHS) and the European Helios project. These two projects were perused for practical and economical reasons. Yet, there is significant cost for developing such standards. More, there are unwanted side effects that need to be minimized. Examples of these side effects are decreasing flexibility and decreasing ownership. (Nielsen 1989 p)

One benefit of having healthcare design guidelines is to speed up the development cycle, and still ending up with high quality interfaces. A significant part of the design decisions will have been set by the guidelines developers. This does mean that these developers carry a huge responsibility.

At least for now, UK’s National Health Services (NHS) is going with the guidelines route and mandating the compliance to these guides for systems to be used by NHS. It will be interesting to see how things will unfold.

Nielsen, J., 1989. Coordinating User Interfaces for Consistency, Academic Pr.

EMRs’ use cases can be broadly classified into data entry and data retrieval. Being a clinician who is frustrated with informaticians overly focus on data entry, I will only present data retrieval use cases.

There are only two use cases:

1. gaining an overview of a patient.

2. searching for specific data.

Nygren, Johnson and Henriksoon (1992) presented these two use cases based on their research of how physicians read medical records. They present a third use case: hypotheses testing. This third use case can be included under ‘searching for specific data’ use case.

Nygren, E. & Henriksson, P., Reading the medical record. I. Analysis of physicians’ ways of reading the medical record. Computer Methods and Programs in Biomedicine, 39(1-2), 1-12.

In my three previous posts titled : ‘we need consistent EMR interfaces because we perceive information in a slow process and fast process’, ‘more on the need for consistency of EMRs’, and  ‘there is no implicit knowledge, but there is implicit knowing’, I presented two reasons for the need for designing consistent interfaces that are pattern rich and where care to the smallest details is made. In this post I will present Nygren, Wyatt and Wright (1998) simplistic view of EMRs interfaces components.

Nygren, Wyatt and Wright (1998) adopted paper medical records components to present these three components of EMRs interfaces:

1. Documents. (or views as called by Microsoft Healthcare Common User Interface guidance)

2. Data contained in these documents

3. Cues

Data as the list of medication a patient is taking can be grouped in a current medication document or view. Cues are intended to draw clinicians attention.

Nygren, E., Wyatt, J.C. & Wright, P., 1998. Helping clinicians to find data and avoid delays. Lancet, 352(9138), 1462-6.

In the previous post ‘how should we define Electronic Medical Records?’, I proposed the adoption of a restrictive definition for EMRs. In the this post I will present Wyatt (1994) view of what data constitute an EMR.

Wyatt (1994) presents a simplistic and yet very practical categorization of data that is stored in EMRs.  This view should hold true with future advancements in medicine. He proposes to modify Weed’s (1968) SOAP clinical data classification. For those not familiar to SOAP classification, S stands for subjective, O for objective, A for assessment and P for plan.

Wyatt’s (1994) classification:

1. Patient findings. This includes Weed’s subjective and objective categories. Wyatt correctly identifies the difficulty of drawing the line between subjective and objective patient data. Patient findings will include patient’s medical history, physical examination findings and investigations’ results.

Read more »

The question I want to pose: what functions should be included in an EMR? And, what functions should not be included?

One of the 100 universal principles of design presented by Lidwell, Holden, and Butler (2003) ,must have book, is the flexibility-usability tradeoff. This principle states as flexibility increases usability decreases. Flexibility is increased by adding functions to an EMR. As functions increase there is much higher chance of this software being harder to develop and harder to maintain. (37signals 2006)

If we agree that medical records must be computerized, then we must agree on a common goal that EMRs will serve. A good definition for an electronic medical record: a computer-based system that captures, stores, or communicates clinical data to enhance medical decision-making. (Wyatt and Crispin 1994) The following three points describe how can this definition be deducted:

Read more »

Tacit knowledge is the type of knowledge we have and use but we find difficulty to express. Explicit knowledge on the other hand, is the type of knowledge we find no difficulty to express. An example of explicit knowledge, is Mr. Smith had diabetes mellitus because his blood sugar is 130mg/dL on two occasions. Tacit knowledge is like when you get a feeling that Mr. Smith is not fully convinced of having diabetes mellitus but you cannot exactly explain why you have this feeling. Tacit knowledge is not available in text books and journal articles but each of us develops it with experience.

A deeper look at Michael Polanyi’s research, points that the above understanding of implicit/explicit knowledge distinction is faulty. Surprisingly, Polanyi who passed away in 1976 is considered the father of implicit/explicit knowledge distinction!

Read more »

Image via Wikipedia

To understand this post you must read my previous first.

The part of our mind that is used for understanding what we read has limited storage capacity. The heart of this part is our working memory. (check this post for description of working memory) I will describe the advantage of this limited capacity in a future post. We need our working memory to be as free as possible to concentrate on our current task. This capacity is sometimes called cognitive load. Automatic information processing does not seem to consume any of our working memory capacity. (Schneider and Chein 2003) This capacity will be fully utilized by controlled information processing, therefore, resulting in better understanding, and probably better decisions.

It is unfortunate that EMRs ignore this dual information processing. These EMRs take things that we used to perceive without effort and force us to use effort to realize them and take cognitive capacity off the main task.

Schneider, W. & Chein, J., 2003. Controlled & automatic processing: behavior, theory, and biological mechanisms. Cognitive Science, 27(3), 559, 525.

These two methods of perceiving information work in parallel. The slow, sequential method is the most obvious and is easy to describe. This slow method is what we use to read and understand.  The other way of perceiving information is fast. Even more, we perceive more then one thing at once (in parallel). It is unfortunate that it is believed that this faster way of perception is uncontrollable. You cannot but perceive the color of the text of this blog post. You also perceive the location of this post on the screen without exerting effort. We must design electronic medical records interfaces that allow us to tap into this fast information processing power. But, how?

Read more »